Frequently Asked Questions

Common questions about Flow Myna, answered honestly — including where the answer is "not yet".

Getting started

What data do I need?

At a minimum:

• Timestamps — when something happened
• Activities — what happened
• Identifiers — what it happened to

Anything else you have makes the analysis richer: attributes such as amounts and statuses, who performed an activity, and several object types rather than one.

Do I need to prepare my data first?

No. The upload wizard reads your files and works out the object and event model, and you approve or correct it before anything is imported. Inconsistent formatting, missing values and column names that mean nothing outside your team are all normal.

How long does it take?

From upload to a process map, usually ten to fifteen minutes, most of which is you reading the proposed mapping rather than waiting for anything.

What file formats are supported?

• CSV — one file or several, up to ten files and 10 GB each
• XES — .xes and .xes.gz
• OCEL 2.0 — .jsonocel and OCEL SQLite

XES and OCEL 2.0 already describe their own objects and events, so they skip the mapping step and go straight to a deterministic import summary.

Excel and JSON are on the roadmap.

Data and security

Is my data secure?

Data is encrypted in transit and at rest, and we send only the minimum needed to each processor we rely on. Flow Myna does not hold a SOC 2 certification of its own; several of the services underneath it do. The security page sets out what is true today, which providers are involved and what they hold, rather than summarising it into a claim.

Where is my data stored?

Core data storage is in Amsterdam, Netherlands. Supporting services and their locations are listed on the security page. Flow Myna Ltd is the legal entity.

Can I get my data back out?

Yes, and this is deliberate:

• Whole dataset as OCEL 2.0 — Export OCEL 2.0 on the dataset page produces a standard JSON log with objects, events and their relations, which other process mining tools can read
• Objects — the object explorer has an Export for the table you are looking at
• Charts and analyses — the Bottlenecks, Health and Seasonality pages export CSV, and copilot charts have a copy control for their underlying data

There is no image export for the map or charts.

How long is data retained?

Your data stays until you delete it. There is no retention schedule, archival rule or storage cap in the product; deleting a dataset, project or workspace is how data goes away.

Concepts

What is process mining?

It discovers how a process actually works by reading the event logs your systems already produce, rather than by asking people to describe it. See What is Process Mining?.

What are variants?

The distinct paths objects take through the process. Real processes have many, not one. See Process Variants.

What is object-centric process mining?

Case-based process mining needs one case ID and one linear flow. Real processes involve orders, items, deliveries and invoices at once, related many-to-many. Object-centric mining models those objects and the relations between them, so you can look at the same events through whichever object you care about. See Object-Centric Approach.

Using the platform

How do I share findings?

Filters live in the URL, so a filtered view is a link anyone with access to the project can open. For anything you will discuss more than once, save the filter so it has a name everyone uses. See Collaboration.

Can I compare time periods?

Yes. Use a timeframe filter, and the Compare page to put two cohorts side by side with the differences ranked. Note that "when" has four meanings — whether the object's whole life, any activity, its start or its end falls in the window — and they give different answers. See Filters & Exploration.

How do I find bottlenecks?

• Bottlenecks page — analyse the wait between any two activities, then read the transitions ranked by average wait; click a row to analyse that span
• Health page — which activities repeat on the same object (rework)
• Process Map — red edges are the long waits; the edge panel shows the distribution behind them
• Co-Pilot — ask which step has the longest wait

What can I ask the copilot?

Anything about the process in the project you have open: "show me the ones that took longer than two weeks", "which block reasons are most common?", "how do these compare with everything else?", "are invoices for blocked orders paid late more often?". See Co-Pilot.

Can I set SLA targets?

Yes. Set a target on the Health or Bottlenecks page names a threshold and shows how many objects would breach it. The target is stored as a rule, and Bottlenecks then judges waits against your target rather than against a raw average, with a Targets tile counting how many are over.

Technical

How does the mapping work?

Three agents run in sequence: one reads and parses your files, one works out what the data means, and one expands that into the review screen. You then approve it, correct it directly, or describe what is wrong and press Remap to have them try again. Nothing is imported before you approve. See Uploading Your Data.

Why both a workspace and a project?

The workspace holds the data; a project is a scoped view of it. That separation lets you load data once and analyse it several ways, keep analyses tidy, and give an outsider one project without the rest of the workspace. See Creating Your Workspace.

Can I send data continuously?

Yes. The API accepts events as they happen, so a dataset can be kept current rather than re-uploaded.

Is there an API for agents?

Yes — Flow Myna exposes an MCP server, so an agent can query your process directly. See Agent Access (MCP).

How big can a dataset be?

Uploads are capped at ten files of 10 GB each. Beyond that it depends on the shape of your data more than its size; talk to us if you are planning something large.

Troubleshooting

My upload failed

Check the extension is supported, the file is under 10 GB, and you are uploading no more than ten files at once. If a CSV came out of a spreadsheet, opening it in a text editor is the quickest way to confirm it really is comma-separated with a header row.

The mapping looks wrong

Fix it in Check Mapping, before approving. Renaming, including and excluding fields, and correcting a timestamp format are all direct edits. For anything structural, write what is wrong and press Remap.

The process map is unreadable

Use Common flows first: its slider narrows the map to the most common flows, and because it drops whole objects rather than parts of them, every number stays real. Crop narrows to one stretch of the process. Dropping rare event types in Data Scope is the more permanent version of the same move.

A filter is not doing what I expected

Open the Active Filters panel and read what is actually in force, including any pinned filter at the top — a pinned baseline is easy to forget about. Check whether an exclusion is behaving as you expect: exclusions are null-safe, so objects that never recorded the attribute are kept.

Plans and billing

What plans are available?

See pricing, or talk to us.

Is there a free trial?

No, but you are not asked to pay blind. Every plan starts with onboarding support: we run your first data load with you, and if you have not reached a trusted answer within thirty days of that load, the month is refunded. See how onboarding works.

Can I change plans myself?

Not yet. There is no self-serve billing in the product — no plan to pick, no invoice history — and nothing in a workspace is metered or charged from inside it. Plans are arranged with us directly. Self-serve billing is on the roadmap.

What happens if I exceed a limit?

There are no usage limits enforced in the product. The Usage page shows what a workspace holds — events, datasets, seats, questions asked in the last day — as counts rather than allowances.

Getting help

How do I contact support?

Email support@flowmyna.com, or use the Feedback control in the top bar of the app.

Where should I start?

• Quick Start Guide — the whole path once
• What is Process Mining? — the concepts
• Co-Pilot — the fastest way to learn what your data can answer

Do you offer training?

Yes — onboarding sessions and team training. Ask us.

Feature requests

Can I request a feature?

Please do. The feedback control in the app reaches us, and so does email. Use cases we do not yet support are the most useful thing you can send.

What's on the roadmap?

More file formats, a read-only role, alert delivery for rules, and self-serve billing. See the roadmap for detail, and the changelog for what has already shipped.